kaashmonee's blog

By kaashmonee, history, 21 month(s) ago, In English

I've always wondered why so many CP platforms require that the submission prints the result to stdout instead of specifying a function name, inputs, and return types like Leetcode. Is there a good reason for this?

  • Vote: I like it
  • -15
  • Vote: I do not like it

»
21 month(s) ago, # |
Rev. 2   Vote: I like it +12 Vote: I do not like it

I think it's because i/o problems are easier to set compared to sig. graded problems. The setter would need to create a different grader for each language.

At least on another judge I've done problem-setting on, sig. graded problems require a custom grader for each supported language you want. After that you set the generator args and/or upload the zip data. You also need to state the technical details of the function for each language on the problem statement. This can cause clutter, especially if you plan on supporting a few languages (at least popular ones like python, java, c++). I've also never seen python/java sig. graded problems (although they might exist?).

Meanwhile for a standard i/o problem, you only need to set generator args and/or upload the zip data. Standard i/o problems also automatically support all the languages available on the judge.